One of the most important metabolic actions of insulin is to promote increased glucose uptake and utilization by the body. Impairment of this action of insulin (insulin resistance) has been associated with many common diseases such as diabetes, obesity, hypertension, and dyslipidemias. Currently, the "gold standard" method for measuring insulin sensitivity in vivo is the hyperinsulinemic euglycemic glucose clamp. This is a very labor intensive test that requires continuous infusions of insulin and glucose and multiple blood samples. An alternative method involves mathematical modeling of an intravenous glucose tolerance test (requires 30 blood samples over 3 hours). The purpose of this project is to develop a reliable method for measuring insulin sensitivity in vivo that is as accurate as currently available methods but simpler to implement. We discovered that the majority of the information needed to accurately estimate insulin sensitivity is contained in the fasting insulin and glucose levels and defined a Quantitative Insulin-sensitivity Check Index (QUICKI) defined as QUICKI = 1/(log (fasting glucose) + log (fasting insulin)). We performed hyperinsulinemic euglycemic glucose clamps as well as intravenous glucose tolerance tests on normal volunteers and groups of patients with diabetes, hypertension, or obesity and found that QUICKI was much more highly correlated with the glucose clamp than minimal model estimates of insulin sensitivity. Thus, we have devised a simpler method for determining insulin sensitivity in vivo that is suitable for studying large populations.